mindspore-Ecosystem
/
mindspore
21
0
Fork 27
Code Issues Pull Requests 6 Projects Releases Wiki Activity

!46061 fix bug: reduce_sum axis to input[1] in clip_by_norm_fission 

Merge pull request !46061 from Yanzhi_YI/master
This commit is contained in:
i-robot 2022-12-02 09:26:16 +00:00 committed by Gitee
parent 2acf7eaa47 43832f4ad5
commit a685b6bca2
No known key found for this signature in database
GPG Key ID: 173E9B9CA92EEF8F
7 changed files with 31 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
mindspore/ccsrc/backend/common/optimizer/common_backend_optimization.cc
View File

@ -33,7 +33,6 @@
#include "backend/common/pass/add_akg_kernel_attrs.h"
#include "backend/common/pass/inplace_assign_for_custom_op.h"
#include "backend/common/pass/flatten_concat_fission.h"
#include "backend/common/pass/clip_by_norm_fission.h"
#include "backend/common/pass/add_dropout_attrs.h"
#include "backend/common/optimizer/dynamic_shape/convert_custom_op.h"
#include "backend/common/optimizer/dynamic_shape/link_custom_op.h"

1
mindspore/ccsrc/plugin/device/ascend/optimizer/ascend_backend_optimization.cc
View File

@ -105,7 +105,6 @@
#include "backend/common/pass/getitem_tuple.h"
#include "backend/common/pass/optimize_dependence.h"
#include "backend/common/pass/erase_visit_attr.h"
#include "backend/common/pass/clip_by_norm_fission.h"
#include "plugin/device/ascend/optimizer/format_type/insert_cast.h"
#include "plugin/device/ascend/optimizer/format_type/convert_unsupported_transnode_to_aicpu.h"
#include "backend/common/pass/eliminate_redundant_op.h"

2
mindspore/ccsrc/plugin/device/gpu/hal/hardware/gpu_device_context.cc
View File

@ -38,7 +38,7 @@
#include "common/graph_kernel/graph_kernel_flags.h"
#include "plugin/device/gpu/hal/profiler/gpu_profiling.h"
#include "plugin/device/gpu/hal/profiler/gpu_profiling_utils.h"
#include "backend/common/pass/clip_by_norm_fission.h"
#include "plugin/device/gpu/optimizer/clip_by_norm_fission.h"
#include "backend/common/session/kernel_graph.h"
#include "plugin/device/gpu/kernel/gpu_kernel.h"
#include "plugin/device/gpu/kernel/gpu_kernel_factory.h"

1
mindspore/ccsrc/plugin/device/gpu/hal/hardware/gpu_session.cc
View File

@ -49,6 +49,7 @@
#include "plugin/device/gpu/optimizer/add_relu_grad_v2_fusion.h"
#include "plugin/device/gpu/optimizer/matmul_biasadd_fusion.h"
#include "plugin/device/gpu/optimizer/neighbor_exchange_v2_fusion.h"
#include "plugin/device/gpu/optimizer/clip_by_norm_fission.h"
#ifdef ENABLE_GPU_INFER
#include "plugin/device/gpu/optimizer/trt_pass/graph_converter.h"
#endif

1
mindspore/ccsrc/plugin/device/gpu/hal/hardware/optimizer.h
View File

@ -56,6 +56,7 @@
#include "plugin/device/gpu/optimizer/insert_cast_gpu.h"
#include "plugin/device/gpu/optimizer/neighbor_exchange_v2_fusion.h"
#include "plugin/device/gpu/optimizer/bias_dropout_add_fusion.h"
#include "plugin/device/gpu/optimizer/clip_by_norm_fission.h"
#include "backend/common/pass/insert_type_transform_op.h"
#endif // MINDSPORE_CCSRC_RUNTIME_HARDWARE_GPU_OPTIMIZER_H_

34
mindspore/ccsrc/backend/common/pass/clip_by_norm_fission.cc → mindspore/ccsrc/plugin/device/gpu/optimizer/clip_by_norm_fission.cc
View File

@ -14,7 +14,7 @@
* limitations under the License.
*/
#include "backend/common/pass/clip_by_norm_fission.h"
#include "plugin/device/gpu/optimizer/clip_by_norm_fission.h"
#include <algorithm>
#include "ir/anf.h"
#include "include/common/utils/anfalgo.h"
@ -61,6 +61,23 @@ std::vector<int64_t> InferBroadcastShape(const std::vector<int64_t> &x_shape, co
}
return broadcast_shape;
}
AnfNodePtr ConvertValueToTensor(const KernelGraphPtr &kernel_graph, const ValueNodePtr &input_node) {
MS_EXCEPTION_IF_NULL(input_node);
auto value_node = input_node->cast<ValueNodePtr>();
MS_EXCEPTION_IF_NULL(value_node);
auto value = value_node->value();
MS_EXCEPTION_IF_NULL(value);
tensor::TensorPtr tensor_ptr = CreateTupleTensor(value->cast<ValueTuplePtr>());
MS_EXCEPTION_IF_NULL(tensor_ptr);
auto tensor_input = std::make_shared<ValueNode>(tensor_ptr);
MS_EXCEPTION_IF_NULL(tensor_input);
tensor_input->set_abstract(tensor_ptr->ToAbstract());
tensor_input = kernel_graph->NewValueNode(tensor_input);
kernel_graph->AddValueNodeToGraph(tensor_input);
tensor_input->set_scope(input_node->scope());
return tensor_input;
}
} // namespace
AnfNodePtr ClipByNormFission::CreateCNodeBase(const FuncGraphPtr &func_graph, const std::vector<AnfNodePtr> &inps,
@ -91,15 +108,14 @@ AnfNodePtr ClipByNormFission::CreateSquareNode(const FuncGraphPtr &func_graph, c
AnfNodePtr ClipByNormFission::CreateReduceSumNode(const FuncGraphPtr &func_graph, const AnfNodePtr &square,
const AnfNodePtr &clip_by_norm, const ShapeVector &shape_vec,
const TypeId &type_id) const {
auto reduce_sum = CreateCNodeBase(func_graph, {square}, kReduceSumOpName, square);
MS_EXCEPTION_IF_NULL(reduce_sum);
auto kernel_graph = func_graph->cast<std::shared_ptr<session::KernelGraph>>();
MS_EXCEPTION_IF_NULL(kernel_graph);
// Sync the attribute of `ClipByNorm` to `ReduceSum`
auto clip_by_norm_prim = common::AnfAlgo::GetCNodePrimitive(clip_by_norm);
MS_EXCEPTION_IF_NULL(clip_by_norm_prim);
auto axis_value = clip_by_norm_prim->GetAttr(kAttrAxis);
MS_EXCEPTION_IF_NULL(axis_value);
common::AnfAlgo::SetNodeAttr(kAttrAxis, axis_value, reduce_sum);
common::AnfAlgo::SetNodeAttr(kAttrKeepDims, MakeValue(true), reduce_sum);
// Get `axis` vector
const auto dim = shape_vec.size();
std::vector<int64_t> axis;
@ -116,7 +132,6 @@ AnfNodePtr ClipByNormFission::CreateReduceSumNode(const FuncGraphPtr &func_graph
MS_EXCEPTION(TypeError) << "For `" << prim::kPrimClipByNorm->name()
<< "`, the type of attribute `axis` is invalid.";
}
// Set abstract to `reduce_sum` op
int64_t ddim = SizeToLong(dim);
ShapeVector reduce_sum_output_shape = shape_vec;
for (const auto &idx : axis) {
@ -127,7 +142,14 @@ AnfNodePtr ClipByNormFission::CreateReduceSumNode(const FuncGraphPtr &func_graph
auto positive_idx = idx < 0 ? idx + ddim : idx;
reduce_sum_output_shape[LongToUlong(positive_idx)] = 1;
}
AnfNodePtr reduce_sum = nullptr;
auto axis_node = NewValueNode(MakeValue<std::vector<int64_t>>(axis));
auto axis_tensor = ConvertValueToTensor(kernel_graph, axis_node);
reduce_sum = CreateCNodeBase(func_graph, {square, axis_tensor}, kReduceSumOpName, square);
MS_EXCEPTION_IF_NULL(reduce_sum);
common::AnfAlgo::SetNodeAttr(kAttrKeepDims, MakeValue(true), reduce_sum);
auto abs = std::make_shared<abstract::AbstractTensor>(TypeIdToType(type_id), reduce_sum_output_shape);
reduce_sum->set_abstract(abs);
return reduce_sum;

0
mindspore/ccsrc/backend/common/pass/clip_by_norm_fission.h → mindspore/ccsrc/plugin/device/gpu/optimizer/clip_by_norm_fission.h
View File